|Hydroxyethyl starch for cardiovascular surgery: a systematic review of randomized controlled trials
|Shi XY, Zou Z, He XY, Xu HT, Yuan HB, Liu H
This review concluded that, compared with control solutions, perioperative hydroxyethyl starch was comparatively safe in cardiovascular surgery. Data was insufficient to make reliable direct comparisons between different preparations, but hydroxyethyl starch 130kD seemed to cause less blood loss and need for transfusions. As review data came from small trials of unclear quality, the authors' conclusions should be treated with caution.
To evaluate the safety of hydroxyethyl starch used in cardiovascular surgery, and to assess which hydroxyethyl starch preparation was the most acceptable.
PubMed (1980 to December 2010), EMBASE (1974 to December 2010) and Cochrane Central Register of Controlled Trials (CENTRAL, Issue 12, 2010) were searched. Search terms were reported. Reference lists of potentially relevant studies were checked. Unpublished studies sought by checking the websites of the US Food and Drug Administration, European Medicines Agency and Fresenius KABI, Germany.
Randomised controlled trials (RCTs) of patients undergoing cardiovascular surgery where hydroxyethyl starch was compared with controls were sought. Eligible trials had to report on blood loss. No restrictions were placed on the type of hydroxyethyl starch administration regime. The primary outcome was blood loss; other outcomes included the need for transfusion, reoperation, renal function, mortality and other complications (pruritus, jaundice, pneumonia, pulmonary oedema, cerebral infarction, heart failure).
The included trials comprised both adults (mean age 44 to 83 years) and children (mean age under one to six years). The interventions used were different hydroxyethyl starch preparations (120 kilo-Dalton (kD), 130kD, 200kD, 264kD, 450kD, and 400kD). Comparators (controls) were albumin, crystalloid, dextran, gelatin, plasma protein fraction, fresh frozen plasma, saline or different hydroxyethyl starch preparations. Some trials included more than two comparison groups. The timing of administration was as priming fluid, volume expander, or both. Methods of measuring blood loss and criteria for blood transfusion varied between trials and (where possible) were reported in the paper.
Two reviewers independently selected studies for inclusion; results were checked by a third reviewer.
Assessment of study quality
Quality was assessed independently by two reviewers using a scoring system based on reporting of items on methods of randomisation, allocation concealment, comparability of baseline characteristics, blinding, drop-outs/withdrawals, and clarity of inclusion criteria. The maximum score was 10. Those trials scoring 5 or below were considered low quality.
Data were extracted to calculate mean differences and 95% confidence intervals (CIs) for continuous data, and risk ratios (RR) with 95% confidence intervals for dichotomous data. Authors were contacted for additional information where necessary.
Data were extracted by two reviewers independently. Disagreements were resolved by consulting a third reviewer.
Methods of synthesis
Each of the hydroxyethyl starch preparations was compared separately with each control solution. Standardised mean differences (SMD) with 95% confidence intervals, and pooled risk ratios with 95% confidence intervals, were calculated using a fixed-effect model. If tests for heterogeneity were significant, a random-effects model was used. Heterogeneity was assessed using Χ² and Ι².
Sensitivity analyses were undertaken by excluding those trials that scored 5 or below for quality, and excluding those that had 15 participants or fewer per group. Subgroup analyses investigated the method of administration (priming, volume expander, both together).
Funnel plots were used to assess publication bias.
Results of the review
Fifty-two RCTs (3,234 participants) were included. Trial size ranged from 15 to 235 participants. Twenty-eight trials had fewer than 50 participants. Six trials scored 10 for quality, four trials scored 9, 13 trials scored 8, 10 trials scored 7, six trials scored 6, 11 trials scored 5, and two trials scored 3.
Compared with albumin, there was a reduction in blood loss with hydroxyethyl starch 130kD (SMD -0.54, 95% CI -0.99 to -0.09; random-effects; seven trials), an increase in blood loss with hydroxyethyl starch 450kD (SMD 0.47, 95% CI 0.26 to 0.68; Ι²=23%; seven trials) and hydroxyethyl starch 400kD (SMD 0.87, 95% CI 0.11 to 1.6; one trial), and no statistically significant differences with hydroxyethyl starch 264kD (Ι²=0%; three trials) or hydroxyethyl starch 200kD (Ι²=47%; seven trials).
Compared with gelatin, one trial showed a statistically significant increase in blood loss with hydroxyethyl starch 450kD, but no differences with hydroxyethyl starch 200kD (seven trials) or hydroxyethyl starch 130kD (10 trials).
Compared with crystalloid, there were no statistically significant differences in blood loss with any of the hydroxyethyl starch preparations (results reported in paper).
In sensitivity analyses when low-quality trials were excluded, the results for blood loss with hydroxyethyl starch 200KD were similar to the main analyses, as were the results for the comparison between hydroxyethyl starch 450kD and albumin. When smaller trials were excluded, there was more blood loss with hydroxyethyl starch 200kD compared with gelatin (SMD 0.29, 95% CI 0.01 to 0.56) and less blood loss compared with crystalloid (SMD -0.65, 95% CI -1.26 to -0.03).
There was no statistically significant difference in number of patients who received blood transfusions with hydroxyethyl starch 200kD compared with albumin (four trials), gelatin (four trials), crystalloid (four trials) or dextran (one trial). Compared with albumin, hydroxyethyl starch 450kD was associated with an increased need for blood transfusion (RR 1.77, 95% CI 1.15 to 2.73; three trials).
When comparing hydroxyethyl starch to controls there was no statistically significant difference in mortality (12 trials, Ι²=0%), renal failure (17 trials), reoperation (13 trials) or complications.
There was insufficient data related to hydroxyethyl starch 400 kD, hydroxyethyl starch 264 kD and hydroxyethyl starch 120 kD to undertake any analyses.
Visual inspection of funnel plots showed asymmetry, indicating publication bias for some outcomes.
Compared with controls, perioperative administration of hydroxyethyl starch preparations was comparatively safe in cardiovascular surgery. Data were insufficient to make reliable direct comparisons between different preparations, but hydroxyethyl starch 130kD seemed to cause less blood loss and transfusions and was suitable for use in cardiovascular surgery.
The aims of this review were clearly stated for inclusion criteria. The search covered a number of relevant sources, including searching for unpublished studies. This reduced the likelihood of publication bias, but the authors reported that tests showed some evidence of it. It was not clear whether language restrictions were applied, so it was difficult to assess the possibility of language bias. The methods of the review were aimed at reducing reviewer error or bias.
The quality of the included trials was assessed using a checklist; since only composite scores were reported, it was difficult to assess the reliability of the evidence presented. Additionally, scoring appeared to be based on reporting of items, rather than potential effects (such as reporting of drop-outs, but not percentage who had dropped out). The methods of synthesis were appropriate. Heterogeneity between trials was investigated.
Much of the data came from small trials with unclear quality. Individual comparisons were generally on data including a relatively small number of participants. As the authors stated, direct comparisons between different preparations of hydroxyethyl starch were not possible. Given these factors, the authors' conclusions should be treated with some caution.
Implications of the review for practice and research
Practice: The authors stated that hydroxyethyl starch 130kD was suitable for use in cardiovascular surgery.
Research: The authors did not state any implications for research.
National Nature Science Foundation of China.
Shi XY, Zou Z, He XY, Xu HT, Yuan HB, Liu H. Hydroxyethyl starch for cardiovascular surgery: a systematic review of randomized controlled trials. European Journal of Clinical Pharmacology 2011; 67(8): 767-782
Subject indexing assigned by NLM
Adult; Aged; Aged, 80 and over; Blood Loss, Surgical /prevention & control; Blood Transfusion; Cardiovascular Surgical Procedures /adverse effects /mortality; Child; Child, Preschool; Hemostasis, Surgical /methods; Hetastarch /adverse effects /chemistry /therapeutic use; Humans; Infant; Middle Aged; Molecular Weight; Perioperative Care; Plasma Substitutes /adverse effects /therapeutic use; Randomized Controlled Trials as Topic
Database entry date
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.